Death and ICU admission
Among pertinent studies, there was insignificant association between mortality and ACEIs/ARBs (OR: 0.97; 95%CI: 0.75 1.27), ACEIs (OR:1.05; 95%CI: 0.75, 1.46), or ARBs (OR:1.18, 95%CI: 0.98, 1.42) (Figure 2; Table 3), regardless of the studies’ country, quality, peer-review status or crude/adjusted measure of effect (Supplementary file 2; Table 4). Similarly, there was an insignificant association between ICU admission and ACEIs/ARBs (OR: 1.09; 95%: 0.65, 1.81) and ACEIs (OR:0.95; 95%CI: 0.65, 1.38) but significantly higher odds of ICU admission with ARBs (OR:1.49, 95%CI: 1.13, 1.97) (Figure 3; Table 3). However, sub-group analyses indicated different results. A significantly lower ICU admission rate was associated with ACEIs/ARBs among European studies (OR:0.49; 95%CI: 0.25, 0.97), and good quality studies (OR:0.36; 95%CI: 0.22, 0.59), in contrast to significantly higher ICU admission rate among USA studies (OR:1.59; 95%CI: 1.28, 1.98), peer-reviewed studies (OR:1.56; 95%CI: 1.23, 1.97), and poor quality studies (OR:1.44; 95%CI: 1.13, 1.84) (Supplementary file 3; Table 4). Meta-analysis of the three studies that reported death and ICU admission as a composite endpoint indicated had significantly lower odds of death/ICU admission with ACEIs/ARBs use (OR:0.67; 95%CI: 0.52, 0.86) but insignificant lower association with ACEIs (OR:0.89; 95%CI: 0.69, 1.14) or ARBs (OR: 0.83; 95%CI: 0.65, 1.06), regardless of any sub-group analysis for ACEIs and ARBs (Figure 4; Table 3). The sub-group analyses for ACEIs/ARBs, however, showed a significantly lower association of death/ICU admission with ACEIs/ARBs only among European studies (OR: 0.68; 95%CI: 0.52, 0.89), good quality studies (OR:0.63; 95%CI: 0.47, 0.84), and studies which reported adjusted measure of effect (OR:0.63; 95%CI: 0.47, 0.84) (Supplementary file 4; Table 4).